Category-based attentional guidance can operate in parallel for multiple target objects

Abstract

The question whether the control of attention during visual search is always feature-based or can also be based on the category of objects remains unresolved. Here, we employed the N2pc component as an on-line marker for target selection processes to compare the efficiency of feature-based and category-based attentional guidance. Two successive displays containing pairs of real-world objects (line drawings of kitchen or clothing items) were separated by a 10 ms SOA. In Experiment 1, target objects were defined by their category. In Experiment 2, one specific visual object served as target (exemplar-based search). On different trials, targets appeared either in one or in both displays, and participants had to report the number of targets (one or two). Target N2pc components were larger and emerged earlier during exemplar-based search than during category-based search, demonstrating the superior efficiency of feature-based attentional guidance. On trials where target objects appeared in both displays, both targets elicited N2pc components that overlapped in time, suggesting that attention was allocated in parallel to these target objects. Critically, this was the case not only in the exemplar-based task, but also when targets were defined by their category. These results demonstrate that attention can be guided by object categories, and that this type of category-based attentional control can operate concurrently for multiple target objects.

Category‐based attentional capture can be influenced by color‐ and shape‐dimensions independently in the conjunction search task

“…The results of N2pc components elicited by distractors demonstrate that category-based attentional capture can operate at relatively early attentional orienting stages. Additional analysis found that the summed contributions of two partially matching distractors were equal to the contributions of fully matching target, indicating the simultaneously and independently orienting operations of two cACSs, in line with previous findings of Nako, Grubert, et al (2016), Wu et al (2016), Jenkins et al (2018). However, the amplitude of N2pc was smaller than previous studies, shows that the allocation of attention according to the guidance of categorical information is less effective than the guidance of physical features, no matter the type of category is semantic-based (Nako, Grubert, et al, 2016;Wu et al, 2013Wu et al, , 2016 or prototype-based (Yang & Zelinsky, 2009; the present study).…”

“…Finally, the two ACSs integrate into a holistic ACS to interfere with consolidation of the target in working memory (evident in the CDA components) and the overall target identification performance (evident by the ACC and RTs). The observation of simultaneous category‐based attentional capture confirms that categorical information, act as high‐level attributes, can be activated at an early stage and then guide attention toward the relevant stimulus (Jenkins, Grubert, & Eimer, ; Wu et al, ).…”

Object features reinstated from episodic memory guide attentional selection

“…In a similar vein, Hout and Goldinger (2015) showed that targets corresponding exactly to the previewed search cue (e.g., a red coffee mug) were found faster than targets oriented differently or targets with a feature that did not match the search cue (e.g., a blue coffee mug). Thus, imprecise or inaccurate features in the attentional template decrease search efficiency (see also Jenkins, Grubert, & Eimer, 2018;Malcolm & Henderson, 2009;Nako, Wu, Smith, & Eimer, 2014).…”

Multiple Correlated Component Analysis for Identifying the Bilateral Location of Target in Visual Search Tasks

“…The N2pc component usually serves as an indicator for attentional selection in the visual search paradigm [20]. N2pc was evoked at about 250 ms after stimulus onset over posterior electrodes contralateral to the visual hemifield of the targets' location in the visual search scene, which was related to the shift of attention towards a candidate target [21]- [23].…”